It is proposed to investigate whether synthesis of poly(adenosine diphosphoribose) acts as a signal to induce G2-arrest in response to radiation-induced DNA damage. This hyphothesis will be tested in G2-phase Chinese hamster ovary (CHO) cells in vitro, using irradiation or 125IUdR incorporation to induce G2-arrest. Mitotic cell selection will be used to monitor cell progression, a permeabilized cell assay to measure poly(adenosine diphosphoribose) polymerase activity, and filter elution to measure DNA damage. Correlations will be sought between poly(adenosime diphosphoribose) polymerase activity, the duration of G2-arrest and DNA damage, as functions of radiation dose. Known inhibitors of poly(adenosine diphosphoribose) polymerase and their inactive analogues will be used over a range of concentrations to perturb expression of these endpoints. The timing, magnitude and direction of these perturbations will be examined to determine whether causal relationships exist. Preliminary data indicate the existence of an additional, non-poly(adenosine diphosphoribose) requiring step in the induction of G2-arrest by radiation, which is susceptible to cordycepin. The effects of combined drug treatments (polymerase inhibitors and cordycepin) will therefore be tested on G2-arrest induced by radiation and on polymerase activity with the purpose establishing if: 1) Agents believed to act solely through poly(adenosine diphosphoribose) inhibition do indeed share a common pathway, 2) Caffeine and related methyl xanthines affect both pathways, 3) The cordycepin susceptible step occurs before, after or in parallel with the poly(adenosine diphosphoribose) requiring step. For this purpose iso-effect data will be interpreted with the aid of drug uptake and competition binding studies. The association of a DNA damage recognition process with radiation-induced G2-arrest will help to provide an understanding, not only of molecular repair processes relevant to survival, but also of the factors governing the rate and probability of cell progression to mitosis. This knowledge may ultimately allow repair processes and cell progression to be manipulated for the construction of improved radiation therapeutic protocols, especially hyperfractionation.